Compliance review of Estimates of additional housing need

Overview

At the time of this report, Welsh Government’s estimates of additional housing need are published as official statistics.

Welsh Government analysts have published estimates of additional housing need since 2019. The estimates are an evidence-based assessment of how many additional homes are likely to be required to ensure that all households have access to adequate and suitable housing which is affordable for them.

They present projected estimates of overall housing need (newly arising and existing unmet need) at the national level, by tenure. The estimates were developed to support the Welsh Government’s National Development Framework, inform ongoing housing policy decision making and have previously been used to support assessments of local well-being under the Well-being of Future Generations (Wales) Act 2015.

This review makes recommendations to strengthen the compliance of the estimates with the Code of Practice for Statistics.

Why we did this review

We became aware of these estimates during our compliance review of Welsh Government household projections. During this review, we noted good links between the projections and the estimates produced by the housing needs team in the Welsh Government.

The latest, 2025-based estimates of additional housing need are published as official statistics, which is a first for these estimates. Previous estimates of additional housing need were not classed as official statistics because they were still being developed.

Given the public interest in these estimates, we have reviewed them to help the statistics team identify opportunities to further improve their adherence to the Code. We have engaged with the statistics team as part of the review process to highlight some areas which would further strengthen adherence with the Code in the future.

Highlighted findings

The development of these housing need estimates as official statistics under the guidance of the Chief Statistician of the Welsh Government reflects the leadership demonstrated by Welsh Government statisticians in promoting a culture that supports analytical integrity, evidence-based decision making and adherence to the Code’s core principles. A Housing Need and Demand Technical Group supported the development of the methods used to produce the estimates. The group included analytical and policy experts who advised on methods and the suitability of the proposed data sources.

The estimates are presented clearly and released in an orderly and transparent way: the publication is publicly pre-announced and access to the final estimates is managed in line with pre-release access legislation. It is not clear when the next estimates of additional housing need will be published. However, we understand that this is based on the timing of the availability of source data, which is not currently clear.

The Welsh Government publishes various documents to support the publication of its official statistics as a ‘statement of compliance’, which includes documents on its approaches to managing quality, confidentiality, release practices and user engagement. We understand that it is committed to updating its documentation as part of its alignment with Code 3.0.

Key limitations are highlighted in the bulletin to convey specific uncertainties and limitations, including an up-front presentation of what the estimates do and do not show. Additional guidance supports the appropriate interpretation of the estimates – for example, higher, principal and lower variant estimates are presented for each measure. This reflects the uncertainties inherent in the underlying household projections. Several annexes in the 2019-based report helpfully break down the methods, step by step. A range of sensitivity tests are presented to show the impact of small changes to the criteria for different variables on the affordable/housing market split for the principal estimate of newly arising need.

The Chief Statistician of the Welsh Government published a blog alongside the 2025-based estimates to communicate the key messages and how they should and should not be used. The team also held a technical press briefing on the day of publication to support the appropriate understanding and use of the estimates by the media. They also held an event for housing sector charities and representative bodies to support their understanding of the estimates.

Changes were made to data sources, assumptions and presentation for the 2025-based estimates. These include publishing a national estimate rather than estimates for the four Welsh regions; presenting existing unmet need as a stand-alone estimate rather than spreading the total over the first five years of the projection; and using new income and private rent prices data sources. The team told us that presenting existing unmet need as a snapshot aimed to improve clarity in how the figures were communicated following user feedback and avoided making assumptions about the pace of clearing unmet need in later years. Publishing at the national level also meant the team could draw on more robust sources of data. While the estimates are now produced at the national level, regional needs are still supported through separate Local Housing Market Assessment guidance and analysis tools. These are produced using different data and assumptions and should not be compared with the estimates of additional housing need statistics.

Some information is included on the limitations of secondary administrative sources, for example on homelessness and concealed households, and those used to inform the breakdowns by tenure. The estimates also draw on census data for overcrowding, and data regarding homeless households in temporary accommodation. Various data sources for affordability, income and growth inform the tenure split and increases. To further support user understanding of strengths and limitations, the strengths and limitations of these data sources for their use in the estimates should be set out more clearly from a Quality Assurance of Administrative Data (QAAD) perspective.

Our judgement

These estimates were developed in line with the Code through collaboration between analytical and policy experts in Welsh Government, and their presentation is of a high standard. We commend the team for their leadership in taking a unique, open and transparent approach to developing these estimates as new official statistics and their efforts to ensure their appropriate interpretation and use. The Welsh Government should continue to progress its plans for the implementation of the new Code of Practice for Statistics.

We have identified three main areas where the Trustworthiness, Quality and Value of these estimates can be strengthened in line with the Code:

The Welsh Government should publish its plans for future estimates of additional housing need to ensure that users are aware of changes before they take place and can feed into future developments.
The team should provide further assurances about the quality of the administrative data sources used and communicate details of the quality checks undertaken, to further support the interpretation of the estimates by users. The team might find OSR’s Quality Assurance of Administrative Data (QAAD) toolkit useful to refer to.
The team should work across the four UK nations to demonstrate how enhanced collaboration between statisticians and policy teams can support the publication of accessible official statistics on housing need. Collaboration with other statistics producers in this way would support the design of consistent and coherent housing need statistics, and better support public awareness of the relative strengths and limitations of housing need estimates.

Next steps

We would like to see the team in Welsh Government continue to improve these statistics and further strengthen their application of the Code. We will continue to engage with the analytical team responsible as they implement these recommendations before the next set of estimates are published.